This project aims to implement a machine learning model to detect possible anomalies in base station 3GPP ETSI KPIs within the 4G LTE radio architecture. It was done by analyzing the statistical data on the network performance that each 4G tower generates every 60 minutes in the form of time series. This study uses machine learning to identify anomaly patterns in different performance evaluation metrics of the 4G network. In addition, the root cause detection of anomalies is also incorporated into the model, pointing out network event counters potentially associated with the network.
- Supervised machine learning (binary classification)
- Feature importance extraction (as root cause)
- Exploratory Data Analysis (EDA)
- Time series (3GPP ETSI LTE RAN KPIs) dataset
- Anomaly detection
- 3GPP ETSI LTE RAN Topology and Optimization
- Network Engineering